Biochemistry 1999, 38, 15853-15862
15853
Essential Thiol Requirement To Restore Pterin- or Substrate-Binding Capability and To Regenerate Native Enzyme-Type High-Spin Heme Spectra in the Escherichia coli-Expressed Tetrahydrobiopterin-Free Oxygenase Domain of Neuronal Nitric Oxide Synthase† Masanori Sono,*,‡ Amy P. Ledbetter,‡ Kirk McMillan,§ Linda J. Roman,| Thomas M. Shea,| Bettie Sue Siler Masters,| and John H. Dawson*,‡,⊥ Department of Chemistry and Biochemistry and School of Medicine, UniVersity of South Carolina, Columbia, South Carolina 29208, Department of Biochemistry, The UniVersity of Texas Health Science Center, San Antonio, Texas 78284-7760, and Pharmacopeia, Inc., Cranbury, New Jersey 08512 ReceiVed July 8, 1999; ReVised Manuscript ReceiVed September 24, 1999
ABSTRACT: Nitric oxide (NO) synthases (NOS) are thiolate-ligated heme-, tetrahydrobiopterin (BH4)-, and flavin-containing monooxygenases which catalyze the NADPH-dependent conversion of L-arginine (L-Arg) to NO and citrulline. NOS consists of two domians: an N-terminal oxygenase (heme- and BH4bound) domain and a C-terminal reductase (FMN- and FAD-bound) domain. In this study, we have spectroscopically examined the binding of L-Arg and BH4 to the dimeric, BH4-free ferric neuronal NOS (nNOS) oxygenase domain expressed in Escherichia coli separately from the reductase domain. Addition of L-Arg or its analogue inhibitors (NG-methyl-L-Arg, NG-nitro-L-Arg) and BH4, together with dithiothreitol (DTT), to the pterin-free ferric low-spin oxygenase domain (λmax: 419, 538, 568 nm) and incubation for 2-3 days at 4 °C converted the domain to a native enzyme-type, predominantly high-spin state (λmax: ∼395, ∼512, ∼650 nm). 7,8-Dihydrobiopterin and other thiols (e.g., β-mercaptoethanol, cysteine, and glutathione, with less effectiveness) can replace BH4 and DTT, respectively. The UV-visible absorption spectrum of L-Arg-bound ferric full-length nNOS, which exhibits a relatively intense band at ∼650 nm ( ) 7.5-8 mM-1 cm-1) due to the presence of a neutral flavin semiquinone, can then be quantitatively reconstructed by combining the spectra of equimolar amounts of the oxygenase and reductase domains. Of particular note, the heme spin-state conversion does not occur in the absence of a thiol even after prolonged (35-48 h) incubation of the oxygenase domain with BH4 and/or L-Arg under anaerobic conditions. Thus, DTT (or other thiols) plays a significant role(s) beyond keeping BH4 in its reduced form, in restoring the pterin- and/or substrate-binding capability of the E. coli-expressed, BH4-free, dimeric nNOS oxygenase domain. Our results in combination with recently available X-ray crystallography and site-directed mutagenesis data suggest that the observed DTT effects arise from the involvement of an intersubunit disulfide bond or its rearrangement in the NOS dimer.
The inorganic free radical gas nitric oxide (NO)1 functions as a neurotransmitter in the brain cerebellum and as a blood flow-controlling factor in endothelial cells. NO is also generated as a cytotoxic agent in macrophages. NO is synthesized in vivo from L-Arg and O2 by a family of † This research was supported by National Institutes of Health Grants GM-26730 (to J.H.D.) and HL-30050 and GM-52419 (to B.S.S.M.), from the United States Public Health Service, and by Grant AQ-1192 from the Robert A. Welch Foundation (to B.S.S.M.). A portion of this paper will be submitted by A.P.L. in partial fulfillment of the Ph.D. requirements at the University of South Carolina. * To whom correspondence should be addressed at the Department of Chemistry and Biochemistry, University of South Carolina, Columbia, SC 29208. Phone: 803-777-5726 (M.S.) or 803-777-7234 (J.H.D.). Fax: 803-777-9521. E-mail:
[email protected] (M.S.) or dawson@ psc.sc.edu (J.H.D.). ‡ Department of Chemistry and Biochemistry, University of South Carolina. § Pharmacopeia, Inc. | The University of Texas Health Science Center at San Antonio. ⊥ School of Medicine, University of South Carolina.
spectrally P450-like, but flavin (FMN and FAD)- and tetrahydrobiopterin (BH4)-containing heme proteins called nitric oxide synthases (NOS) through NADPH-dependent monooxygenation reactions (for selected reviews, see refs 1-5). Three isoforms of NOS are known which have considerable amino acid sequence homology (50-60%) to each other, but differ in certain aspects. Brain neuronal (constitutive) 1 Abbreviations: NO, nitric oxide; P450, cytochrome P450; BH , 4 (6R)-5,6,7,8-tetrahydrobiopterin; BH2, 7,8-dihydrobiopterin; 4-aminoBH4, 2,4-diamino-5,6,7,8-tetrahydro-6-(L-erythro-1,2-dihydroxypropyl)pteridine; NOS, nitric oxide synthase; “nNOS, iNOS, and eNOS”, brain neuronal, macrophage (inducible), and endothelial nitric oxide synthases, respectively; CaM, calmodulin; P450BM-3, fatty acid-hydroxylating cytochrome P450 isolated from Bacillus megaterium; P450-CAM, camphor-hydroxylating cytochrome P450 isolated from Pseudomonas putida; DTT, dithiothreitol; BME, β-mercaptoethanol; EDTA, ethylenediaminetetraacetic acid; Tris, tris(hydroxymethyl)aminomethane; NNA, NG-nitro-L-Arginine; NMA, NG-methyl-L-Arginine; MCD, magnetic circular dichroism.
10.1021/bi991580j CCC: $18.00 © 1999 American Chemical Society Published on Web 11/05/1999
15854 Biochemistry, Vol. 38, No. 48, 1999 NOS (nNOS) has ∼220 N-terminal amino acid extra residues (containing the so-called “GLGF motif”) that probably serve as an intracellular anchor domain for cellular protein-protein interactions, resulting in cellular localization of this isoform (6). Thus, nNOS (∼160 kDa) is considerably larger than the endothelial (constitutive) (eNOS) and macrophage (inducible) (iNOS) (∼130 kDa for the latter two) isoforms. Among the three isoforms, only eNOS is membrane-associated through myristoylation of its N-terminal residues; nNOS and iNOS are cytosolic. NOS is only catalytically active in its homodimeric form (7-9). The C-terminal half of NOS consists of the reductase domain (∼80 kDa) which contains the NADPH-binding site and protein-bound FAD and FMN. The latter two prosthetic groups serve as an electron-transport chain from NADPH to the N-terminal half, the oxygenase (heme-bound) domain (1-5). The two domains are connected by a calmodulin (CaM)-binding module. The binding of Ca2+-CaM serves as a switch for interdomain electron transfer to initiate catalysis (10). Thus, NOS is a catalytically self-sufficient monooxygenase similar to P450BM-3 from Bacillus megaterium (11) except that the latter does not require BH4 and Ca2+-CaM. The NOS reductase domain shares striking sequence homology with mammalian NADPH-P450 reductase, as well as with the reductase domain of P450BM-3. In contrast, the oxygenase domain of nNOS has a sequence that distinctly differs from those of P450s, thereby excluding NOS from the cytochrome P450 gene family. Recent X-ray crystal structures of the BH4- and L-Arg-bound dimeric iNOS and eNOS oxygenase domains (12-15) have, in fact, revealed structural distinctions between NOS and P450 as well. Among the four NOS prosthetic groups, FAD, FMN, heme, and BH4, the first three components serve as redox cofactors. The exact role(s) of BH4 in NOS catalysis remain(s) to be clarified. Previous studies have indicated that BH4 is required to stabilize the catalytically active dimeric structure of NOS (16, 17). Yet, more recent studies have also implicated BH4 in redox reactions in the NOS catalysis either as a two-electron-transferring cofactor (18, 19) as in aromatic amino acid hydroxylases (20, 21) or as a unique one-electron-transferring redox cofactor (13, 22). Recently, the intact (full-length) as well as the separated oxygenase and reductase domains of the three recombinant NOS isoforms have been successfully expressed in E. coli (8, 2326). Since E. coli lacks the BH4 synthesis system (27), this expression technique has enabled researchers to prepare completely BH4-free NOS proteins. This has made it possible to study the effects of BH4 on the spectral, structural, and catalytic properties of intact NOS, as well as its separated oxygenase domain. The E. coli-expressed, BH4- and substrate (L-Arg)-free oxygenase (heme-bound) domain of rat nNOS exhibits a largely low-spin-type electronic absorption spectrum (23), in sharp contrast to the predominantly high-spin-type spectral properties of BH4-bound, Arg-free native full-length nNOS (28, 29). In the present study, we have examined the binding of BH4 as well as substrate or inhibitors to the pterin-free, dimeric nNOS oxygenase domain. We have been able to convert the low-spin oxygenase domain to a native enzymetype high-spin form by addition of L-Arg and BH4 together with DTT. The characteristic electronic absorption spectrum of full-length ferric nNOS containing bound BH4 and L-Arg,
Sono et al. which is somewhat unusual because of the presence of a neutral semiquinone as compared with the spectra of other thiolate-ligated heme-containing proteins (29), has then been successfully reconstructed from both the oxygenase and reductase domains.2 Interestingly, however, we have observed that the heme spin-state conversion, i.e., BH4 and/or L-Arg binding to the domain, does not occur without a thiol even under anaerobic conditions where BH4 is kept in its fully reduced form. The significance of this finding is discussed in relation to an intersubunit disulfide bond (12, 15) or intersubunit Zn(Cys)4 metal center (13-15) which has been recently revealed by X-ray crystallography for the dimeric iNOS and eNOS oxygenase domains. EXPERIMENTAL PROCEDURES Materials. BH4, 4-amino-BH4, and BH2 were purchased from Research Biochemicals, BIOMOL Research Laboratories, and Dr. B. Schirck’s laboratory (Jona, Switzerland), respectively. NG-Nitro-L-Arg (NNA) and NG-methyl-L-Arg (NMA) were purchased from Alexis Biochemicals. CO and argon gases were obtained from Matheson Co. All other chemicals including L-Arg and L-thiocitrulline were purchased from Sigma or Aldrich and used as received. Preparation of Full-Length nNOS and the Separated nNOS Oxygenase (Heme-Bound) and Reductase (FlaVin-Bound) Domains. E. coli-expressed full-length rat brain nNOS and its separately expressed, heme-containing oxygenase domain (the N-terminal 714 amino acid residues) and the flavincontaining reductase domain (amino acid residues 7151429) were prepared as described previously (23, 24). Concentrated full-length nNOS protein and the oxygenase domain (∼100 µM on the basis of heme) were stored at -70 °C in 50 mM Tris-HCl (pH 7.4) containing 10% glycerol, 1 mM BME, and 150 mM NaCl. The stock solutions of the reductase domain (∼120 µM) were also stored in the same buffer without BME and NaCl. Spectral and Substrate-, Inhibitor-, and Heme LigandBinding Experiments for the nNOS Oxygenase Domain. All experiments in this study were carried out at 4 °C to minimize spectral changes of the proteins associated with their denaturation during prolonged incubation periods (up to 10 days). Stock solutions of L-Arg (500 mM), NNA (20 mM), NMA (20 mM), thiocitrulline (20 mM), DTT (0.1 and 1 M), and BME (1 M) were prepared in water, and those of BH4 and 4-amino-BH4 (10 mM) were made in water containing either 0.1 M DTT or 0.1 M BME. BH2 was dissolved in dimethyl sulfoxide as a 20 mM stock solution. All of these solutions were stored in small aliquots (100250 µL) at -70 °C. For anaerobic incubation experiments with the nNOS oxygenase domain, a small aliquot (25-30 µL) of concentrated protein stock solutions was diluted in argon-bubbled buffer (∼300 µL) in a rubber septum-sealed cuvette (0.2 cm) by using a gastight microsyringe. While continuously purging oxygen by blowing argon into the cuvette, desired volumes (400 nm initially drastically decreased due to reduction of the flavins. The solution was then gently bubbled with ∼20 µL of air using a microsyringe. The absorbance of the resulting reductase domain between 400 and 700 nm gradually increased with time (in ∼20 min), as in the case of P450 reductase (30), to give a relatively stable neutral semiquinone spectrum having absorption bands at 590 and 630 nm, in addition to those at 456 and 476 nm. Determination of Subunit-Based Protein Concentrations. The concentrations of full-length nNOS and the nNOS oxygenase domain were determined based on (1 heme/ subunit protein) the pyridine hemochromogen assay (556 ) 34 mM-1 cm-1) (29). The concentration of ferricyanideoxidized reductase domain (23) was determined based on 2 flavins/subunit protein using 445 ) 20.6 mM-1 cm-1 at pH 7.4 (30-32). Optical Absorption and Magnetic Circular Dichroism (MCD) Spectroscopy. Electronic absorption spectra were recorded with a Varian/Cary 210 or 219 spectrophotometer interfaced to IBM PCs. MCD spectra were recorded with a Jasco J500A spectropolarimeter equipped with a Jasco MCD1B electromagnet operated at 1.41 T. All spectral measurements were performed at ∼4 °C using 0.1, 0.2, 0.5, or 1 cm quartz cuvettes (see ref 29). RESULTS Spectral Properties and Complex Formation of the nNOS Oxygenase Domain with Substrate, Inhibitors, and Heme Ligands. The nNOS oxygenase domain used in this study was judged to be dimeric based on a gel filtration chromatogram at its final purification step.3 The MCD and electronic absorption spectra of the BH4-free ferric oxygenase domain
Biochemistry, Vol. 38, No. 48, 1999 15855
FIGURE 1: MCD (A) and electronic absorption spectra (B) of the BH4-free ferric nNOS oxygenase domain and camphor-free ferric P450-CAM. The nNOS oxygenase domain in the absence (s) and presence of 10 mM L-Arg (‚ ‚ ‚) and camphor-free P450-CAM (- - -). The spectra in the visible region (480-700 nm) are enlarged 7.5 times. The spectra of the nNOS oxygenase domain (10.8 µM) were recorded at 4 °C in 50 mM potassium phosphate buffer (pH 6.5) containing 10% (v/v) glycerol and 0.1 mM EDTA using 0.5 cm (MCD) and 0.2 cm (UV-visible) cuvettes. The spectra of P450CAM are taken from ref 33.
(the N-terminal amino acid residues 1-714) in the absence and presence of 10 mM L-Arg are shown in Figure 1, panels A and B, respectively, together with the spectra of camphorfree ferric P450-CAM (33). Both proteins exhibit very similar low-spin-type absorption and MCD spectra with respect to band positions and intensities, and spectral patterns. Given the fingerprinting capability of MCD spectroscopy (29, 34), this indicates that the ferric oxygenase domain most likely has an [RS--FeIII(heme)-OH2] active site coordination sphere structure, where RS- ) Cys--415 for nNOS (23, 35), as does camphor-free P450-CAM (34, 36). No detectable spectral differences were observed for the oxygenase domain in potassium phosphate buffer (pH 6.5 and 7.5) and Tris buffer (pH 7.5). Addition of the substrate L-Arg (10 mM) (Figure 1, dotted lines) or substrate analogue inhibitors (1 mM, NNA and NMA) (not shown) to the ferric oxygenase domain (Figure 1, solid lines) caused only very small spectral changes, if any, even after overnight incubation at 4 °C. The ferric oxygenase domain forms low-spin complexes with heme ligands such as imidazoles (unsubstituted and substituted), cyanide, and DTT. These oxygenase domain complexes exhibit absorption and MCD spectra (not shown) very similar to those of the corresponding P450-CAM derivatives (33, 37, 38) in band positions, intensities, and spectral patterns. Similar MCD spectra (except for some differences in intensity) for the corresponding complexes 3 The nNOS oxygenase domain expressed in E. coli and used in this study was primarily dimeric as judged by the final gel sizing column chromatogram (Hiload 26/60 Superdex 200 gel, Pharmacia). Shoulder fractions containing the monomer were excluded.
15856 Biochemistry, Vol. 38, No. 48, 1999
Sono et al.
Table 1: Electronic Absorption Spectral Parameters and Dissociation Constants (Kd Values) for Heme Ligand Complexes of the Ferric nNOS Oxygenase Domain at 4 °C λnm (mM-1 cm-1) Soret heme
liganda
none Imid N-MeImid cyanide DTT
Kd (mM)
visible
δ
γ
β
R
360 (32) 361 (35) 361 (36) 368 (40) 376 (56)
419 (100) 429 (92) 429 (92) 439 (69) 460 (57)
538 (11.9) 545 (12.3) 545 (12.1) 556 (11.2) 560 (13.6)
568 (11.2) 570d (9.4) 570d (9.3)
pH
6.5b
0.13 1.7 1.1
pH 7.5c 0.022 0.038 0.12
a Imid, imidazole; N-MeImid, N-methylimidazole; DTT, dithiothreitol. b 50 mM potassium phosphate containing 10% glycerol and 0.1 mM EDTA. c 50 mM Tris containing 10% glycerol and 0.1 mM EDTA. d Shoulder.
with imidazole and cyanide for full-length eNOS have also been recently reported (39). Absorption spectral data as well as dissociation equilibrium constants (Kd values) of four selected complexes are summarized in Table 1. All the imidazole derivatives examined in this study formed lowspin complexes with the ferric nNOS oxygenase domain which exhibit similar absorption spectra to one another. The observed pH dependence of the Kd values for the Nmethylimidazole (pKa ) 7.25, see ref 40 for imidazoles) and DTT (pKa ) 8.1) complexes indicates that it is the unprotonated (neutral) form of N-methylimidazole and the deprotonated (anionic) form of DTT that coordinate to the ferric heme iron of the enzyme. Similar results have been previously reported for DTT binding to camphor-free P450-CAM (38). The Kd values of the complexes at pH 7.5 are similar for unsubstituted imidazole (pKa ) 6.9), N-phenylimidazole (pKa ) ∼5.85, estimated) (Kd ) ∼20 µM for both), and N-methylimidazole (Kd ) 38 µM). However, the affinity for 4-phenylimidazole (pKa ) 6.0) (Kd ) 0.69 mM) is considerably lower (by >15-fold) than those for the others. Interestingly, while the 4-phenylimidazole complex is low-spin for the nNOS oxygenase domain, the analogous complex is highspin for full-length eNOS (39) and nNOS (this study). The Kd value ()0.12 mM) of the DTT-oxygenase domain complex at pH 7.5 was either not affected (Kd ) 0.12 mM) by 10 mM L-Arg or 1 mM NMA or somewhat affected (Kd ) 0.25 mM) by 1 mM NNA. Spectral Changes Due to Spin-State ConVersion upon Binding of L-Arg, Inhibitors, and BH4 to the Ferric nNOS Oxygenase Domain. Binding of L-Arg, its analogue inhibitors, or BH4 to NOS has been shown to correlate with spectral (spin state) changes of the ferric enzyme by using radioisotopelabeled compounds for binding and spectrophotometric techniques for the spin-state changes (17, 24, 41, 42).4 Individual addition of either the substrate L-Arg (10 mM) or its analogue inhibitors NNA and NMA (1 mM), or BH4 (0.25 mM), to the low-spin ferric oxygenase domain did not cause readily detectable spin-state conversion of the heme iron even after incubation for 2 days at 4 °C under aerobic conditions. However, when BH4 (0.25 mM) was added together with DTT (2.5 mM) either in the presence or in the absence of L-Arg, NNA, or NMA, noticeable, although quite 4 Recent crystal structural analysis of the BH - and/or L-Arg-bound 4 dimeric oxygenase domains of iNOS and eNOS indicates that there is a close linkage, through an extensive hydrogen bond network, between pterin/substrate binding and formation of the five-coordinate (highspin) heme (see refs 12-15 for details). Thus, it is unlikely that BH4 and/or L-Arg could bind to the low-spin NOS oxygenase domain without perturbing the spectrum of the heme in the oxygenase domain.
FIGURE 2: (A) Electronic absorption spectral changes of the ferric nNOS oxygenase domain (16.9 µM) from a low-spin to high-spin type after addition of 2 mM L-Arg, 0.25 mM BH4, and 2.5 mM DTT as a function of incubation time. The spectra were recorded using a 0.5 cm cuvette after incubation for 0 h (‚ ‚ ‚), 24 h (- ‚ -), 42 h (- - -), and 70 h (s). See the legend to Figure 1 for other conditions. (B) Normalized electronic absorption spectra of the incubated (70 h) nNOS oxygenase domain as described in panel A before (s) and after (‚ ‚ ‚) subtraction of a fractional (15%) spectrum of the DTT complex (- - -). The spectrum of the dithionitereduced ferrous-CO form (- ‚ -) of the same incubated nNOS oxygenase domain sample is also shown.
slow, spectral changes were observed during prolonged incubation at 4 °C. Figure 2A displays representative results where, upon concomitant addition of L-Arg (2 mM) and BH4 (0.25 mM) together with DTT (2.5 mM) at pH 6.5, the lowspin-type absorption spectrum of the oxygenase domain (see Figure 1B, solid line) was initially quickly (80%) dimeric in their BH4-free forms (8, 13, 41, 53) in these cases. Requirement of DTT for Binding of BH4 and L-Arg to NOS and for Catalytic Function of NOS. The most striking finding in this study is that neither BH4 nor L-Arg readily binds to the nNOS oxygenase domain unless a thiol (e.g., DTT) at 2.5-10 mM is present. Low concentrations (0.1-0.2 mM) of BME that had been brought in from the stock solutions of the nNOS oxygenase domain had little effect. Moreover, only in the presence of DTT, BH2 (a BH4 autoxidation product which is not reducible with DTT) can also bind to the oxygenase domain (Table 2). Stuehr, Mayer, and their co-workers have reported that, in the case of the pterin-free iNOS oxygenase domain, BH2 can bind to the protein in competition with BH4 with somewhat lower affinity than BH4 (42). More recently, Stuehr’s group has demonstrated that BH2 and several other pterin derivatives can induce spinstate conversion of full-length iNOS as well as its separated oxygenase domain to similar extents to those caused by BH4 (53), despite the fact that BH2 cannot support the NOS catalysis (53). Again their buffer contained 1 mM DTT or 2.5 mM BME. Based on these results, we propose that DTT (and other thiols), in addition to maintaining the pterin in the reduced form (i.e., BH4), plays a crucial role(s) in converting the BH4-free nNOS oxygenase domain from a form which is inaccessible to BH4 and/or L-Arg to an accessible form. A thiol (DTT or BME in 1-2.5 mM) is always included in a standard NOS assay mixture. Many thiols (also including glutathione and cysteine) are known to dramatically enhance the catalytic activity of purified NOS by up to severalfold in dose- and preincubation time-dependent manners (52, 54,
Sono et al. 55). These effects of thiols are explained by (a) their ability to maintain BH4 in its catalytically active reduced form (5456) and, in part, by (b) possible reductive protection of the NOS protein (52, 55). However, it should be emphasized that these previous studies have not specifically examined the requirement of a thiol for binding of BH4 and/or L-Arg to either BH4-free full-length NOS or BH4-free NOS oxygenase domain as we have done in this study for the nNOS oxygenase domain. The present study clearly rules out the BH4-regenerating role of DTT for its requirement. Since the nNOS oxygenase domain used in this study was judged to be dimeric vide supra,3 DTT does not appear to be required simply for dimerization of the domain for the binding of the pterin and/or substrate. However, the thiol requirement for BH4 and/or L-Arg binding as well as the significant binding rate acceleration by preincubation with DTT are suggestive of protein cysteine group involvement in the restoration of the pterin- and substrate-binding capability of the separated domain. Relevant to this proposal, recent crystallographic as well as cysteine residue point mutation studies by several research groups are described below. Recently, Masters, Poulos, and co-workers (13) and, shortly afterward, Fischmann and co-workers (14) have determined the X-ray crystal structures of the dimeric eNOS and iNOS (and also nNOS6) oxygenase domains. The eNOS (and nNOS6) oxygenase domain used by the former group was prepared by trypsin cleavage of the E. coli-expressed full-length enzyme (13). The structures for both isoforms revealed the presence of a Zn(Cys)4 metal center at the dimer interface which involves two conserved cysteine groups (Cys101 and Cys96 for eNOS) and is likely important for the stability of the protein dimeric structure as well as the binding of BH4 and L-Arg to the protein (13, 14). It should be pointed out that both of these research groups crystallized the BH4- and L-Arg-bound oxygenase domains under reducing conditions using tris(2-carboxyethyl)phosphine (13), glutathione sulfonate (13), or DTT (14). When zinc is absent, one of the thiol ligands (Cys101 in bovine eNOS, Cys99 in human eNOS, Cys109 in murine iNOS, and Cys331 in rat brain nNOS) is proposed to form an intersubunit disulfide bond (13). Crane, Stuehr, and co-workers (12) first detected such a disulfide bond in the murine iNOS oxygenase domain dimer structure. In fact, Poulos and collaborators have quite recently demonstrated that either the disulfide bond-containing or the Zn(Cys)4 metal center-containing (with added ZnSO4 and a reducing agent) structure can be prepared for the human iNOS oxygenase domain under different crystallization conditions (15). Other laboratories had shown earlier that the Cys99Ala mutant of human eNOS (57) and the Cys109Ala mutant of murine iNOS (42) had partially impaired BH4-binding capabilities. As isolated, these mutants gained ∼20% and ∼70% of the full catalytic activity of the wild-type enzymes, respectively, when sufficiently high concentrations (>100 µM) of BH4 were added. Further, Masters and co-workers have recently demonstrated that the full-length Cys331Ala nNOS mutant is largely low-spin and defective in L-Arg binding and catalysis (58). However, prolonged (∼12h at 4 °C) incubation of the mutant in the presence of a relatively 6 C. S. Raman, H. Li, T. L. Poulos, P. Marta ´ sek, and B. S. S. Masters, unpublished observations.
Thiol-Restored BH4/Arg Binding to nNOS Heme Domain high concentration of L-Arg (10 mM), with or without BH4 (300 µM), has restored the high-spin-type absorption spectrum. More significantly, the full catalytic activity (with 10 µM BH4) was also restored for the L-Arg-incubated mutant even in the absence of the enzyme-bound zinc (58, 59). The Cys331Ala nNOS mutant lacks Zn-binding capability (59). Thus, Zn is not essential for catalysis. In view of these recent findings, the thiol-assisted restoration of the pterin- and substrate-binding capability of the E. coli-expressed nNOS oxygenase domain might be related to the reductive cleavage or rearrangement of the intersubunit disulfide bond. The bond most likely involves either one or both of the two conserved cysteine groups (Cys331 and Cys326 for nNOS), as detected by X-ray crystallography for the homologous residue in iNOS (Cys109) (12, 15). In the present case, zinc does not appear to be playing a crucial role; the zinc content in the E. coli-expressed nNOS oxygenase domain preparation has been shown to be less than stoichiometric (i.e., 0.2 mol/mol of dimer ) 0.1 mol/ mol of heme) (59). In addition, any zinc that might be inadvertently contained in the buffer used would be unavailable for the oxygenase domain because EDTA (0.1 mM), which is added to the buffer, would remove the free zinc. We have recently found that the DTT-assisted restoration of the BH4- and substrate-binding capability of the nNOS oxygenase domain is essential for generation of a stable dioxygen complex of the dithionite-reduced nNOS oxygenase domain in 50% (v/v) ethylene glycol at -30 °C (60). In conclusion, this study has demonstrated that a thiol, DTT in particular, is essential for the binding of BH4 and L-Arg to the E. coli-expressed pterin-free ferric nNOS oxygenase domain. Thus, in the separate oxygenase domain, we have been able to generate a heme iron center that is spectroscopically identical to the full-length ferric nNOS active site structure having a five-coordinate high-spin ferric heme. This work will facilitate future studies aimed at achieving a better understanding of the significance of the unique cysteine-involving dimeric conformation of the NOS oxygenase domain and its active site structure, which have been recently revealed by X-ray crystallography (12-15). ACKNOWLEDGMENT We thank Drs. Edmund W. Svastits and John J. Rux for developing the computer-based spectroscopic data-handling system and Dr. John W. Ledbetter (Medical University of South Carolina) for his generous gift of BH2. REFERENCES 1. Marletta, M. A. (1994) Cell 78, 927-930. 2. Bredt, D. S., and Snyder, S. H. (1994) Annu. ReV. Biochem. 63, 175-195. 3. Griffith, O. W., and Stuehr, D. J. (1995) Annu. ReV. Physiol. 57, 707-736. 4. Moncada, S., and Higgs, E. A. (1995) FASEB J. 10, 13191330. 5. Masters, B. S. S., McMillan, K., Sheta, E. A., Nishimura, J. S., Roman, L. J., and Martasek, P. (1996) FASEB J. 10, 552558. 6. Brenman, J. E., Chao, D. S., Gee, S. H., McGee, A. W., Craven, S. E., Santillano, D. R., Wu, Z., Huang, F., Xia, H., Peters, M. F., Froehner, S. C., and Bredt, D. S. (1996) Cell 84, 757-767. 7. Baek, K. J., Thiel, B. A., Lucas, S., and Stuehr, D. J. (1993) J. Biol. Chem. 268, 21120-21129.
Biochemistry, Vol. 38, No. 48, 1999 15861 8. Rodrı´guez-Crespo, Gerber, N. C., and Ortiz de Montellano, P. R. (1996) J. Biol. Chem. 271, 11462-11467. 9. Klatt, P., Pfeiffer, S., List, B., Lehner, D., Glatter, O., Ba¨chinger, H. P., Werner, E. R., Schmidt, K., and Mayer, B. (1996) J. Biol. Chem. 271, 7336-7342. 10. Abu-Soud, H. M., and Stuehr, D. J. (1993) Proc. Natl. Acad. Sci. U.S.A. 90, 10769-10772. 11. Narhi, L. O., and Fulco, A. J. (1987) J. Biol. Chem. 262, 6683-6690. 12. Crane, B. R., Arvai, A. S., Ghosh, D. K., Wu, C., Getzoff, E. D., Stuehr, D. J., and Tainer, J. A. (1998) Science 279, 21212126. 13. Raman, C. S., Li, H., Marta´sek, P., Kral, V., Masters, B. S. S., and Poulos, T. L. (1998) Cell 95, 939-950. 14. Fishmann, T. O., Hruza, A., Niu, X. D., Fossetta, J. D., Lunn, C. A., Dolphin, E., Prongay, A. J., Reichert, P., Lundell, D. J., Narula, S. K., and Weber, P. C. (1999) Nat. Struct. Biol. 6, 233-242. 15. Li, H., Raman, C. S., Glaser, C. B., Blasko, E., Young, T. A., Parkinson, J. F., Whitlow, M., and Poulos, T. L. (1999) J. Biol. Chem. 274, 21276-21284. 16. Klatt, P., Schmidt, K., Lehner, D., Glatter, O., Ba¨chinger, H. S., and Mayer, B. (1995) EMBO J. 14, 3687-3696. 17. Mayer, B., Wu, C., Gorren, A. C. F., Pfeiffer, S., Schmidt, K., Clark, P., Stuehr, D., and Werner, E. R. (1997) Biochemistry 36, 8422-8427. 18. Rusche, K. M., Spiering, M. M., and Marletta, M. A. (1998) Biochemistry 37, 15503-15512. 19. Perry, J. M., and Marletta, M. A. (1998) Proc. Natl. Acad. Sci. U.S.A. 95, 11101-11106. 20. Kaufman, S. (1997) Tetrahydrobiopterin: Basic Biochemistry and Role in Human Disease, The Johns Hopkins University Press, Baltimore, MD. 21. Kappock, T. J., and Caradonna, J. P. (1996) Chem. ReV. 96, 2659-2756. 22. Bec, N., Gorren, A. C. F., Voelder, C., Mayer, B., and Lange, R. (1998) J. Biol. Chem. 273, 13502-13508. 23. McMillan, K., and Masters, B. S. S. (1995) Biochemistry 34, 3686-3693. 24. Roman, L. J., Sheta, E. A., Marta´sek, P., Gross, S. S., Liu, Q., and Masters, B. S. S. (1995) Proc. Natl. Acad. Sci. U.S.A. 92, 8428-8452. 25. Marta´sek, P., Liu, Q., Liu, J., Roman, L. J., Gross, S. S., Sessa, W. C., and Masters, B. S. S. (1996) Biochem. Biophys. Res. Commun. 219, 359-365. 26. Wu, C., Zhang, J., Abu-Soud, H., Ghosh, D. K., and Stuehr, D. J. (1996) Biochem. Biophys. Res. Commun. 222, 439-444. 27. Tho¨ny, B., Leimbacher, W., Bu¨rgisser, D., and Heizmann, C. W. (1992) Biochem. Biophys. Res. Commun. 189, 1437-1443. 28. Stuehr, D. J., and Ikeda-Saito, M. (1992) J. Biol. Chem. 267, 20547-20550. 29. Sono, M., Stuehr, D. J., Ikeda-Saito, M., and Dawson, J. H. (1995) J. Biol. Chem. 270, 19943-19948. 30. Iyanagi, T., and Mason, H. S. (1973) Biochemistry 12, 22972308. 31. Masters, B. S. S., Kamin, H., Gibson, Q. H., and Williams, C. H., Jr. (1965) J. Biol. Chem. 240, 921-931. 32. Vermilion, J. L., and Coon, M. J. (1978) J. Biol. Chem. 253, 2694-2704. 33. Dawson, J. H., Andersson, L. A., and Sono, M. (1982) J. Biol. Chem. 257, 3606-3617. 34. Dawson, J. H., and Sono, M. (1987) Chem. ReV. 87, 12551276. 35. Richards, M. K., Clague, M. J., and Marletta, M. A. (1996) Biochemistry 35, 7772-7780. 36. Poulos, T. L., Finzel, B. C., and Howard, A. J. (1987) J. Mol. Biol. 195, 687-700. 37. Dawson, J. H., Andersson, L. A., Sono, M., and Hager, L. P. (1992) New J. Chem. 16, 577-582. 38. Sono, M., Andersson, L. A., and Dawson, J. H. (1982) J. Biol. Chem. 257, 8308-8320. 39. Berka, V., Palmer, G., Chen, P.-F., and Tsai, A.-L. (1998) Biochemistry 37, 6136-6144.
15862 Biochemistry, Vol. 38, No. 48, 1999 40. Hoffmann, K. (1953) in Imidazole and Its DeriVatiVes (Part I), pp 13-15, Interscience Publishers, New York. 41. Gorren, A. C. F., List, B. M., Schrammel, A., Pitters, E., Hemmens, B., Werner, E. R., Schmidt, K., and Mayer, B. (1996) Biochemistry 35, 16735-16745. 42. Ghosh, D. K., Wu, C., Pitters, E., Moloney, M., Werner, E. R., Mayer, B., and Stuehr, D. J. (1997) Biochemistry 36, 10609-10619. 43. Matsuoka, A., Stuehr, D. J., Olson, J., Clark, P., and IkedaSaito, M. (1994) J. Biol. Chem. 269, 20335-20339. 44. Frey, C., Narayanan, K., McMillan, K., Spack, L., Gross, S. S., Masters, B. S. S., and Griffith, O. W. (1994) J. Biol. Chem. 269, 26083-26091. 45. Salerno, J. C., McMillan, K., and Masters, B. S. S. (1996) Biochemistry 35, 11839-11845. 46. Schenkman, J. B., Remmer, H., and Estabrook, R. W. (1967) Mol. Pharmacol. 3, 113-123. 47. Kaufman, S. (1959) J. Biol. Chem. 234, 2677-2682. 48. Gachhui, R., Presta, A., Bentley, D. F., Abu-Soud, H. M., McArthur, R., Brudvig, G., Ghosh, D. K., and Stuehr, D. J. (1996) J. Biol. Chem. 271, 20594-20602. 49. Berka, V., Chen, P.-F., and Tsai, A.-L. (1996) J. Biol. Chem. 271, 33293-33300. 50. White, R. E., and Coon, M. J. (1982) J. Biol. Chem. 257, 3073-3083. 51. Abu-Soud, H. M., Loftus, M., and Stuehr, D. J. (1995) Biochemistry 34, 11167-11175.
Sono et al. 52. Gorren, A. C. F., Schrammenl, A., Schmidt, K., and Mayer, B. (1997) Biochemistry 36, 4360-4366. 53. Presta, A., Siddhanta, U., Wu, C., Sennequier, N., Huang, L., Abu-Soud, H. M., Erzurum, S., and Stuehr, D. J. (1998) Biochemistry 37, 298-310. 54. Komori, Y., Hyun, J., Chiang, K., and Fukuto, J. M. (1995) J. Biochem. 117, 923-927. 55. Hofmann, H., and Schmidt, H. H. W. (1995) Biochemistry 34, 13443-13452. 56. Bublitz, C. (1977) Biochem. Med. 17, 13-19. 57. Chen, P.-F., Tsai, A.-L., and Wu, K. K. (1995) Biochem. Biophys. Res. Commun. 215, 1119-1129. 58. Marta´sek, P., Miller, R. T., Liu, Q., Roman, L. J., Salerno, J. C., Migita, C. T., Raman, C. S., Gross, S. S., Ikeda-Saito, M., and Masters, B. S. S. (1998) J. Biol. Chem. 273, 3479934805. 59. Miller, R. T., Marta´sek, P., Raman, C. S., and Masters, B. S. S. (1999) J. Biol. Chem. 274, 14537-14540. 60. Ledbetter, A. P., McMillan, K., Roman, L. J., Masters, B. S. S., Dawson, J. H., and Sono, M. (1999) Biochemistry 38, 8014-8021. 61. Sono, M., McMillan, K., Roman, L. J., Masters, B. S. S., and Dawson, J. H. (1997) J. Inorg. Biochem. 67, 135 (Abstr.). BI991580J
